Produces summary table of non-A occurrences within an analyzed dataset.

Creates a per-transcript summary table with read counts, non-A residue counts and hits, and poly(A) tail length statistics, grouped by user-defined factors (e.g. sample, condition).

Usage

summarize_nonA(
  merged_nonA_tables,
  summary_factors = c("group"),
  transcript_id_column = c("ensembl_transcript_id_short")
)

Arguments

merged_nonA_tables

Data frame or tibble. Output of merge_nonA_tables.

summary_factors

Character string or vector of strings. Column name(s) used for grouping (default: "group").

transcript_id_column

Character string. Column containing the transcript identifier (default: "ensembl_transcript_id_short", as added during data pre-processing; can be changed by the user).

Value

A tibble with one row per transcript per group, containing:

polya_median

Numeric. Median poly(A) tail length for the transcript.

polya_mean

Numeric. Mean poly(A) tail length for the transcript.

counts_total

Integer. Total number of reads mapped to the transcript.

counts_blank

Integer. Number of reads with no non-A residues.

counts_nonA / hits_nonA

Integer. Reads with any non-A / total non-A hits.

counts_C / hits_C

Integer. Reads with C / total C hits.

counts_G / hits_G

Integer. Reads with G / total G hits.

counts_U / hits_U

Integer. Reads with U / total U hits.

Details

The distinction between counts and hits:

• counts — number of reads containing at least one occurrence of a given non-A residue type.

• hits — total number of occurrences of a given non-A residue type across all reads (a single read may contribute multiple hits).

See also

merge_nonA_tables for preparing the input, calculate_fisher for statistical testing on merged data, annotate_with_biomart for adding gene-level annotation.

Examples

if (FALSE) { # \dontrun{

summarized <- ninetails::summarize_nonA(
  merged_nonA_tables = merged_nonA_tables,
  summary_factors = "group",
  transcript_id_column = "ensembl_transcript_id_short")

} # }